New Music Video: Undertow I've been feeling the weight lately; the constant churn of the algorithm, the pressure to keep up, drowning in the noise. So I stopped fighting. I let myself drift for a bit. And somewhere in that, this came through: "What if letting go isn't losing

Family Life

Another fantastic edition of Brian’s weekly Zine! The images this week are 🔥 I’ve been a subscriber for months and always look forward to seeing what Brian has cooked up.

today, sync. is excited to announce react-1 a ten billion parameter masked video diffusion model built for one purpose: giving you full creative control over performances in post

All these image models, and more, are available in @LeonardoAi Do you have a favourite?

6. Seedream 4 Bright and commercial. Dynamic angle, electric blue eyes.

5. GPT-Image-1 Film photography energy. Vintage grade, polo dress interpretation.

4. Nano Banana Pro Went full tennis. Visor, wristbands, clay court; all unprompted extras.

3. Phoenix 1.0 A Leonardo AI foundational model. It interpreted the prompt as a dramatic close-up. Nailed the blue eyes and bold brows.

2. Lucid Realism Softer, editorial. Natural skin, muted tones. The eye colour is a bit darker but Lucid Realism is built to generative diverse images, which can help avoid generic looking people.

1. Lucid Origin Commercial polish. Added Nike branding of its own accord. Very natural eyes with the flecks of colour that make it feel lifelike.

Same prompt. Six genAI models. Wildly different results. I asked Leonardo AI for a blue-eyed woman on a tennis court pulling her hair into a ponytail. Here's what each model gave me 👇 🧵

Ready to reveal an IP developed over the past seven months. Introducing The Ninth Step: Chapter I In 1959, nine Soviet ski hikers died in the northern part of the Ural Mountains ridge under strange circumstances. Only one came back. This is their story. Read more below:

I love Brian's weekly zine. Each episode is full of stunning images with prompts, which you can then put your own twist on. I amended a prompt from this issue, used Lucid Origin in Leonardo then refined using Magnific's Skin Enhancer in Freepik. Zine details in post below.

I premiered my most realistic and cinematic AI short to date at the Bionic awards showcase at the Rich Mix Cinema, which you can watch below. Made with @freepik using the latest tools: @GoogleDeepMind Nano Banana Pro, @Kling_ai, @dreamina_ai Seedream, 2.6 and @Suno. It was

Comments and notifications are broken again for me. A recent post of mine shows I have 3 comments but I can't see any of them. I always try to reply to all comments, so apologies if you've left one and I've not acknowledged it.

Quiet words. @LeonardoAi just added Kling 2.6 Audio and lip sync, no extra steps. (Start frames created in Lucid Realism).

The team @freepik are on fire this week! Who needs sleep, right? I created this video using: Seedream 4.5 (out tomorrow) Magnific Skin Enhancer Kling 2.6 which has baked-in sound/lip sync

I'm testing out Kling's new O1 video model on Leonardo AI. You can use multiple reference images, created in different models and with different aspect ratios, to create a single, coherent video. Explanation below.

You can try Kling O1 on @LeonardoAi now.

I think the model's understanding of the various elements and the coherence of the final video is very impressive. Traffic and pedestrians in the background move correctly (which isn't often the case with earlier models) and the movements of the man and dragon follow the prompt